Discriminative Artificial Intelligence(AI):

Discriminative Artificial Intelligence(AI) models are taught to distinguish between different data classes, patterns, images etc. In simple words it is a type of artificial intelligence that helps us tell different things apart. Discriminative AI focuses on learning the lines that separate different groups in our data. Instead of making new data, it looks at data we already have to figure out what it is.

?Application Areas:

Image Recognition: Discriminative AI helps identify what's in pictures, like telling if it's a cat or a dog.

Speech Recognition: It figures out different words or phrases when people talk.

?Natural Language Processing (NLP): It sorts text into different groups, like figuring out if a review is positive or negative.

How Discriminative AI Works:

Training Data: Discriminative AI learns from examples, like pictures of cats and dogs, to get better at telling them apart.

Learning Boundaries: It figures out the lines that separate different things, and the more examples it sees, the better it gets.

Classification: When given new data, it can tell what it is based on what it learned before.

As technology gets better, discriminative AI will likely get better too, making more accurate decisions. It will become even more important, helping us make smarter choices across different areas.

Generative Artificial Intelligence(AI):

Generative AI means a part of artificial intelligence that's really good at making new stuff.

It can create things like text, images, music, videos, and even code.

It does this by learning from existing data and then making something new based on what it learned.

Discriminative Artificial Intelligence(AI) models are already available in Generative Artificial Intelligence(AI) model, so this model,

• Does Not identify difference.
• Does Not solve problems
• Create new information only

Which Domains deal with unstructured data?

There are three domains that deal with unstructured data,

1. NLP(Natural Language Processing)
2. Computer Vision
3. Speech Recognition

NLP( Natural Language Processing:

NLP is a domain which is interested in textual language. When AI applies to Natural Language or Text, it is called NLP.

CV(Computer Vision):

CV is a domain of AI which is interested in Image or video content. It is a branch of AI in which we study methods to solve videos or image problems.

Speech Recognition:

?It is a branch of AI, which deals with voice such as wavelength, pitch, frequency etc.

?How Does Generative AI Work?

Generative AI uses something called generative models to do its magic. These models learn from lots of data and then make up new things. One famous kind of generative model is the Large Language Model (LLM) . Here's how it works:

• Training: Generative AI learns from a mix of different data.
• For example, it might read millions of sentences from books, articles, and websites.
• Learning Patterns: The model looks at this data to understand how things fit together, like how words make sentences and how different words relate to each other. It also learns the style of writing from different authors.
• Generating New Content: Once it's learned enough, the model can make up new text. So, if you ask it to write a poem about the moon, it'll come up with something poetic and original.

Here is an important things to note that:

• Both NLP and CV can be Discriminative Artificial Intelligence(AI) or Generative Artificial Intelligence(AI)
• Before 2012, these three domains were studied? separately.
• After 2012, these three domains of AI are studied together in Deep Learning.

Examples of generative AI:

Examples of Generative AI in Action:

• Chatbots: Generative AI powers chatbots, enabling them to provide contextually relevant responses to user queries.
• Image Synthesis: Advanced AI systems like DALL-E generate images from textual descriptions, demonstrating the ability to translate words into visual representations.
• Code Generation: Generative AI assists developers by automatically generating code snippets based on given specifications or requirements.
• Music Composition: AI composers like Magenta compose music, leveraging generative AI techniques to create melodies, harmonies, and entire compositions.

How to make a Generative Artificial Intelligence(AI)? Model?

Generative Artificial Intelligence(AI) requires Generative Artificial Intelligence(AI) Model. To make Generative Artificial Intelligence(AI) Model, in start inputs are given to the computer that is called Seed Input. Seed input is the initial point of generating the? random numbers.?

What are two popular models of Generative AI ?

There are two popular models of? Generative AI:

1. LLM (Large Language Model)
2. Diffusion or Latent Model

☆LLM (Large Language Model):

This model is related to language so this exists in the NLP domain. These types of models are made up of some specific Neural Network architecture (having billions of parameters or neurons), which is called GPT(Generative Pretrained Transformer). In other words, we can say that LLM is a specific type of transformer called GPT.

How LLM (Large Language Model) work?

Mainly, two steps are involved in the working of LLM.

Step 1 is a method called Tokenization.?

☆Tokenization:

In the working of LLM, tokenization plays a crucial role. It's a process of breaking down the

text or prompt into smaller units like phrases, words, or even characters. These smaller units,

called tokens, form the vocabulary of the generative AI model. Each GPT model has its own

limits of tokens or vocabulary, beyond which the model won’t be able to generate output.

The limit of input tokens accessed is called the context window of that LLM.

In summary, generative AI opens doors to endless possibilities, blending creativity with

data-driven insights. As technology advances, responsible development and ethical usage

remain paramount in harnessing its full potential.

It is a process of breaking down the text or prompt into smaller units like phrases, words etc.

Tokens can also be called Vocabulary of Generative AI Model. An important point to be considered is, every GPT Model has its own limits of Tokens or vocabulary, above these vocabulary limits, Model won’t be able to generate output.?

Limits of input Token accessed is called Context Window of that LLM.

☆ Step 2 is after Tokenization, GPT Model will be able to process further.

☆ Diffusion or Latent Model:

? A diffusion model is a type of generative model that can create data similar to the data it is trained on, such as images, text, or audio. A diffusion model works by gradually adding noise to the data until it becomes random, and then learning to reverse this process by removing the noise.

Diffusion models are a kind of computer program that can make new things that look like the things they learned from, such as pictures, words, or sounds. Diffusion models work by making the things they learned from more and more messy until they are just random, and then learning how to make them less messy again. Diffusion models can make new things by starting from random mess and making it less messy using what they learned .

☆Examples of Diffusion models:

Diffusion models are very popular these days, because they can do very well in many tasks, such as making new pictures, videos, or molecules. Some examples of diffusion models are DALL-E 2, Stable Diffusion, and Midjourney. Diffusion models have some good points over other computer programs, such as not needing to compete with other programs, being more stable and strong, and being able to handle different kinds of things .

Xeven Solutions

Artificial Inspiration

Artificial Intelligence